mark zenodo data optional

.github/workflows/python-ci.yml

@@ -82,7 +82,7 @@ jobs:
 
     - name: Run all dataset tests
       run: |
-        python -m pytest test -v --cov=brukerapi --cov-branch --cov-report=xml --cov-report=term-missing --cov-report=html
+        python -m pytest test -v --download_test_data --cov=brukerapi --cov-branch --cov-report=xml --cov-report=term-missing --cov-report=html
 
     - name: Upload coverage HTML
       uses: actions/upload-artifact@v7

test/conftest.py

@@ -20,6 +20,7 @@ def pytest_addoption(parser):
     parser.addoption("--test_data", action="store", default="")
     parser.addoption("--test_suites", action="store", default="")
     parser.addoption("--properties_reference", action="store", default="")
+    parser.addoption("--download_test_data", action="store_true", default=False)
 
 
 # -------------------------------
@@ -49,22 +50,29 @@ def pytest_addoption(parser):
 
 
 def pytest_sessionstart(session):
-    for dataset in ZENODO_FILES:
-        _ensure_test_data(dataset)
-    for dataset in GITHUB_DATASETS:
-        _ensure_github_test_data(dataset)
+    if not session.config.getoption("download_test_data"):
+        return
+
+    for dataset in _requested_dataset_names(session.config.getoption("test_data")):
+        if dataset in ZENODO_FILES:
+            _ensure_test_data(dataset)
+        elif dataset in GITHUB_DATASETS:
+            _ensure_github_test_data(dataset)
 
 
 # -------------------------------
 # Helpers
 # -------------------------------
-def _resolve_requested_datasets(opt: str | None):
+def _requested_dataset_names(opt: str | None):
     if not opt or opt.lower() == "all":
-        available_local_datasets = [name for name in LOCAL_DATASETS if (TEST_DATA_ROOT / name).is_dir()]
-        return [*ZENODO_FILES, *GITHUB_DATASETS, *available_local_datasets]
+        return [*ZENODO_FILES, *GITHUB_DATASETS, *LOCAL_DATASETS]
     return [opt]
 
 
+def _resolve_requested_datasets(opt: str | None):
+    return [name for name in _requested_dataset_names(opt) if (TEST_DATA_ROOT / name).is_dir()]
+
+
 def _is_required_github_data_file(path: Path):
     return path.name in {"2dseq", "traj"} or path.name.startswith("rawdata.job")
 

test/test_dataset.py

@@ -10,6 +10,7 @@
 from brukerapi.exceptions import UnsuportedDatasetType
 
 data = 0
+PV51_STUDY_PATH = Path("test/test_data/PV51/0.2H2")
 
 
 def test_unsupported_dataset_type(tmp_path):
@@ -27,6 +28,7 @@ def test_unsupported_dataset_type(tmp_path):
         ("test/test_data/PV51/0.2H2/10/pdata/1", "2dseq"),
     ],
 )
+@pytest.mark.skipif(not PV51_STUDY_PATH.is_dir(), reason="PV51 test data is not available")
 def test_directory_constructor_uses_default_load(path, dataset_type):
     dataset = Dataset(path)
 

test/test_folders.py

@@ -1,8 +1,12 @@
 from pathlib import Path
 
+import pytest
+
 from brukerapi.dataset import Dataset
 from brukerapi.folders import Folder, Processing, Study
 
+PV51_STUDY_PATH = Path("test/test_data/PV51/0.2H2")
+
 
 def test_folder_traversal_skips_processed_spectra(tmp_path):
     experiment_path = tmp_path / "1"
@@ -35,9 +39,10 @@ def test_folder_traversal_skips_processed_spectra(tmp_path):
     assert processing_datasets == {"2dseq"}
 
 
+@pytest.mark.skipif(not PV51_STUDY_PATH.is_dir(), reason="PV51 test data is not available")
 def test_study_get_dataset_returns_fid_and_2dseq():
     study = Study(
-        Path("test/test_data/PV51/0.2H2"),
+        PV51_STUDY_PATH,
         dataset_state={"parameter_files": [], "property_files": [], "load": 0},
     )